adharshramu12/seetha_ai

Seetha AI is a free, open-source desktop assistant for Windows that you control with your voice. You talk to it, it understands what you want, and it carries out actions on your computer for you. The project is built in Python and runs entirely on your own machine, so there is no account to create, no server involved, and no telemetry tracking your usage. You bring your own API key from a provider like NVIDIA, and your data stays local. What makes Seetha different from a typical chatbot is that it can plan and execute multi-step tasks. If you ask it to open Notepad and write a poem, the assistant figures out the steps, performs them, and then takes a screenshot to check that the action actually worked. If something failed, it retries. This observe, act, and verify loop means it can handle real computer automation rather than just returning text. The assistant covers a wide range of everyday tasks. You can ask it to open or close applications, adjust volume and screen brightness, set reminders, send WhatsApp messages, play YouTube videos, or create instant meeting links. It also has a persistent memory, so it remembers facts you share across restarts. A keyboard shortcut brings up a command bar from inside any application, and the assistant includes an animated avatar that lip-syncs while speaking. Seetha is designed to work even without an internet connection. Simple commands like media control and reminders are handled by a local rules engine with no API calls needed. More complex requests require an AI model, and the software can fall back between providers like NVIDIA NIM and Groq if one goes down. The project includes safety guards, such as confirming destructive actions by voice before proceeding. To install it, you clone the repository and run a batch file that sets up the environment automatically. The full README is longer than what was shown.